宁夏和新疆水稻种质资源表型遗传多样性分析及综合评价

为了拓宽宁夏水稻种质资源的遗传基础,本研究对193份宁夏及新疆水稻种质资源的表型性状进行了遗传多样性分析与综合评价,结果表明：大部分水稻种质资源呈现出茎节包、叶片不早衰、剑叶角度小、抗倒性强、穗立形状弯曲、落粒性难;14个数量性状的遗传变异均达到了极显著水平,其中单株产量的变异系数最大,为23.8%;遗传多样性指数变幅为1.86~2.08,平均2.02;通过聚类分析将193份水稻种质资源划分为8个类群,大部分种质资源分布在第Ⅰ类群,占比65.3%,第Ⅱ类群占比9.3%,第Ⅲ类群占比20.7%,第Ⅳ-Ⅷ类群共包含9份种质资源,总占比4.7%。通过主成分分析,将14个数量性状指标转换为7个主成分,累积贡献率为90.67%;利用隶属函数与权重对193份水稻种质进行综合评价,获得了可综合评价种质资源农艺性状的D值,排名前5的种质资源均来自宁夏,表现出较好的综合性状,其中以宁218表现最为突出,新疆种质资源新引区2015-13的D值最低,综合性状表现最差;利用逐步回归分析,获得了株高、穗长、每穗实粒数、结实率、一次枝梗数、粒长宽比、单株产量、单穗粒重、千粒重9个关键表型性状指标。本研究剖析了宁夏及新疆水稻种质资源的亲缘关系,探索了水稻农艺性状的综合评价方法,可为拓宽宁夏水稻种质资源的遗传基础提供理论参考。

Analysis and Comprehensive Evalution of Phenotypic Genetic Diversity of Ningxia and Xinjiang Rice Germplasm

In order to broaden the genetic basic of rice germplasm resources in Ningxia and Xinjiang, the genetic diversity and comprehensive assessment of 193 Ningxia and Xinjiang rice germplasm resources were analyzed based on phenotypic traits. The results indicated that most of these rice germplasm resources presented coated stem node, not premature leaf, smal sword leaf angle, strong lodging resistance, panicle shape bending and difficult seed-shattering. The genetic variation of 14 quantitative traits all reached a very significant level, and the variation coefficient of single plant yield was the biggest, 23.8%. The variation range of genetic diversity index was from 1.86 to 2.08 with average 2.02. By cluster analysis, 193 rice germplasm resources were divided into 8 groups. Most of the germplasm resources were in the first group, accounting for 65.3%. The second group was accounted for 9.3% and the third group was 20.7%. The rest of these groups contained 9 kinds of rice germplasm which was accounted for 4.7%. Through principal component analysis, 14 quantitative traits were converted into 7 principal components, and the cumulative contribution rate was reached 90.67%. An objective comprehensive evaluation value (D-value) of rice agronomic traits was obtained using membership function with index weight method. The five highest of the D value of germplasm resources were all from Ningxia, which showed better comprehensive traits, with the most outstanding performance in Ning218. Nevertheless, Xinyinqu2015-13, which a germplasm resource from Xinjiang showed the worst synthetical properties. Using stepwise regression and principal component analysis, the plant height, spike length, grain number per panicle, seed setting rate, primary branch number, grain aspect ratio, grain yield per plant, single spike grain weight and 1000-grain weight were selected as the key indexes for rice phenotypic traits.